Youth Included in the Report

The youth who are the focus of this report represent only a subset of youth with disabilities who received special education services in secondary school in 2001, not the entire population. The full population to which the NLTS2 sample generalizes is a cohort of youth who were ages 13 through 16 and received special education services in grade 7 or above in participating schools and school districts as of December 1, 2000. Weights for analyses reported in this document are calculated so that all youth who responded for themselves to a telephone interview or completed a mail questionnaire generalize to the subset of that cohort who would be capable of responding for themselves. To illustrate, consider the following groups:

• A = The entire NLTS2 sample.
  
  
• A1 = The portion of A for whom no contact was attempted because parents stated that youth were unable to respond to an interview or complete a questionnaire. This also includes youth known to be deceased.
  
  
• A2 = The portion of A for whom contact was attempted by telephone or mail survey because their parents stated they were capable of responding and, in the case of telephone interviews for youth younger than 18, gave consent for an interview.

For each of these sample groups, there is a corresponding group in the universe, which can be denoted with a "B," such that the universe is B, the portion of the universe whose parents would state that they are unable to respond (had they been included in the sample) is denoted B1, and the remaining portion is denoted B2. The sizes of these universe subgroups can be estimated by weighting all youth in A (as if they all had been respondents) up to the entire universe, B. Then the sum of the weights of all youth in A, A1, and A2 are estimates of the number of youth in B, B1, and B2.

However, not all youth in A2 were interviewed or completed a questionnaire. Let those who did respond be labeled A2r. Weights were computed (adjusting for various youth and school characteristics used as stratifying or post-stratifying variables) that project A2r up to B2. Thus the youth survey weights for A2 respondents project to the portion of the universe (B2) for whom interviews would be attempted if all individuals in the universe had participated in NLTS2.

The subgroup of youth who could respond for themselves differ in several ways from those whose parents indicated they were unable to respond. Appendix B provides detailed information regarding differences between these groups, examples of which are summarized briefly here.

The disability profiles of the group of youth who responded for themselves do not differ significantly from the profiles of those whose parents were interview respondents. Youth respondents are more likely than youth whose parents were respondents to have high self-care skills (96 percent vs. 90 percent, p < .01), but these two groups do not differ significantly in their functional cognitive skills or social skills. Youth respondents also are less likely to have trouble communicating (26 percent vs. 43 percent, p < .001), understanding language (30 percent vs. 43 percent, p < .01), and using their arms and hands for fine motor activities (4 percent vs. 11 percent, p < .01). Consistent with these differences, they also are less likely to have a disability identified in their first year of life (13 percent vs. 25 percent, p < .01), although there is no significant difference in their rate of receiving special education services their first years in school.

Differences in youth's services are apparent. For example, youth who were not their own respondents were more likely to receive a several related and support services (e.g., occupational therapy and transportation services).

No significant demographic differences or differences in youth's instructional programs between the two respondent groups are apparent.

Top